[Angstrom-devel] Ångström going forward
dp at xora.org.uk
Thu May 21 09:49:20 CEST 2009
Ok, there have been discussions on #oe and with some people in private
over the last few days that make me think it is time to discuss how
Ångström works in general and how we interface with OE.
1) Bug tracking, it is becoming clear that non Ångström OE guys are
getting frustrated by the bugs we put into the OE bugtracker. Do we want
to setup our own bugtracker for our own bugs, pushing them to the OE one
when they are actual OE problems. I'm quite happy to oversee this if the
decision is yes, but I'd prefer some software simpler than bugzilla. I
can also arrange hosting on a server that isn't as loaded as l2g.
2) Overlaying/Forking/Other, sometimes what we want to do is different
from what other distros built from OE want to do. This causes friction.
Now Chris 'kergoth' Larson has checked in his work on collections it
should be easy to manage an Ångström overlay which has higher priority
than OE core data so we don't cause so much friction. Recipes/changes
from this overlay can of course be pushed back into OE core.
At this current time I am against forking OE into an Ångström only tree
although this is very tempting as we could reduce the
machines/tasks/images/distros to only approved working ones.
3) Anything else? Yes its time to discuss how Ångström functions from
now on. Its pretty clear that the "core" team idea never really stuck.
But what is the Ångström stucture? At the moment it seems to be a few
developers and a whole lot of users, this is fine, do we even need to
change the way we work.
One thing I would like to see is the re-instatement of machine mentors
for Ångström. Where we only "support" machines that have an active
maintainer. If he drops off the radar, resigns, the machine gets moved
to unsupported until someone else picks up the baton.
Anyway guys this is a community project, lets hear your views.
Graeme 'XorA' Gregory
More information about the Angstrom-distro-devel